April 14, 2012


Tips for how patients can evaluate health research (Connie Midey, Apr. 13, 2012, The Republic)

"It's about having a healthy skepticism," she says. [physician Lisa Schwartz, professor at Dartmouth Medical School and author with Steven Woloshin and H. Gilbert Welch of "Know Your Chances: Understanding Health Statistics"] suggests starting with: "What is this study trying to say, and do I believe it? Is it about people like me, or is it an early result, perhaps from animal research?"

If research concludes that a certain intervention -- a drug or food or treatment -- reduces risk, she would ask, "My risk of what exactly? Is it something I really care about? Does it affect whether I live or die? Or is it just about a blood test or X-ray result?"

The National Institutes of Health lists seven questions to ask about medical findings. Schwartz would add: How big is the expected benefit? What are the side effects? Do the research subjects have the same level of disease you have? If not a randomized, controlled trial, was it an observational study?

The first type is "the gold standard," Schwartz says. "People are randomly assigned to the thing being tested, such as to the new drug or to the old drug. Then when we see differences at the end of the study, we can feel confident that they are because of the intervention we did."

Observational studies have a place, too, sometimes out of necessity. Researchers can't assign subjects to do something harmful, such as smoke cigarettes.

But with a study that follows people over time, "it's hard to separate out one contributing factor from all their lifestyle habits," Schwartz says.

Seven questions to ask when you learn about a new medical finding...

Posted by at April 14, 2012 8:44 AM

blog comments powered by Disqus